1. Field of Invention
This invention relates to a low-volume biomarker generator used in radiopharmaceutical production.
2. Description of the Related Art
Cyclotrons are used to generate high energy charged particle beams for purposes such as nuclear physics research and medical treatments. One area where cyclotrons have found particular utility is in the generation of biomarkers for medical diagnosis by such techniques as positron emission tomography (PET). A conventional cyclotron involves a substantial investment, both in monetary and building resources. In addition to a large size and weight, the power requirements often involve a dedicated and substantial electrical power system due to the high voltage supply necessary for the radio frequency system to accelerate the particles into a beam sufficient to overcome the binding energy (nominally 7-9 MeV) causing a stable target isotope to become a radioisotope. Thus, medical facilities have a need for biomarkers, but the monetary, structural, and power requirements of conventional cyclotrons have historically made it impracticable for most hospitals and other medical facilities to produce biomarkers on-site.
The half-life of clinically important positron-emitting isotopes, i.e., radionuclides, relative to the time required to process a radiochemical is a significant factor in biomarker generation. The large linear dimensions of the reaction vessel in radiochemical synthesis systems commonly used in biomarker generators result in a small ratio of surface area-to-volume and effectively limit the heat transfer and mass transport rates and lengthens processing time. The four primary PET radionuclides, fluorine-18, oxygen-15, nitrogen-13, and carbon-11, are considered to have short half-lives. For example, fluorine-18 has a half-life of approximately 110 minutes. Converting nucleophilic fluorine-18 ([18F]F) into the biomarker [18F]fluorodeoxyglucose ([18F]FDG) requires approximately 45 minutes using one of the larger conventional radiochemical synthesis systems. The processing time is significant with, respect to the half-life of the radioisotope, with a processing time-to-half-life ratio of approximately 40%. Because some of the radioisotope will decay during processing, the percent yield of the biomarker is reduced, in this case, to a range of approximately 50 to 60%. Even with efficient distribution networks, the short half-lives and low yields require production of a greater amount of the biomarker than is actually needed for the intended use.
Recent advancements have led to the development of smaller reaction systems. By reducing the linear dimensions of the reaction vessel used in the radiochemical synthesis system, the ratio of surface area-to-volume and, consequently, heat transfer and mass transport rates increases. The smaller size of the reaction vessels lends itself to replication allowing multiple reaction vessels to be placed in parallel to simultaneously process the biomarker. In addition to faster processing times and reduced space requirements, these smaller reaction systems require less energy. However, such advancement has not been seen with the cyclotrons necessary for radioisotope production.
A biomarker is used to interrogate a biological system and can be created by “tagging” or labeling certain molecules, including biomolecules, with a radioisotope. A biomarker that includes a positron-emitting radioisotope is required for positron emission tomography (PET), a noninvasive diagnostic imaging procedure that is used to assess perfusion or metabolic, biochemical and functional activity in various organ systems of the human body. Because PET is a very sensitive biochemical imaging technology and the early precursors of disease are primarily biochemical in nature, PET can detect many diseases before anatomical changes take place and often before medical symptoms become apparent. PET is similar to other nuclear medicine technologies in which a radiopharmaceutical is injected into a patient to assess metabolic activity in one or more regions of the body. However, PET provides information not available from traditional imaging technologies, such as magnetic resonance imaging (MRI), computed tomography (CT), and ultrasonography, which image the patient's anatomy rather than physiological images. Physiological activity provides a much earlier detection measure for certain forms of disease, cancer in particular, than do anatomical changes over time.
A positron-emitting radioisotope undergoes radioactive decay, whereby its nucleus emits positrons. In human tissue, a positron inevitably travels less than a few millimeters before interacting with an electron, converting the total mass of the positron and the electron into two photons of energy. The photons are displaced at approximately 180 degrees from each other, and can be detected simultaneously as “coincident” photons on opposite sides of the human body. The modern PET scanner detects one or both photons, and computer reconstruction of acquired data permits a visual depiction of the distribution of the isotope, and therefore the tagged molecule, within the organ being imaged.
In the field of nuclear medicine, it is well known that cyclotrons are used for
producing radiopharmaceuticals for use in imaging. Most clinically important positron emitting radioisotopes are produced in a cyclotron. Cyclotrons, including two-pole, four-pole, and eight-pole cyclotrons, operate by accelerating electrically-charged particles along outward, quasi-spherical orbits to a predetermined extraction energy generally on the order of millions of electron volts. The high-energy electrically-charged particles form a continuous beam that travels along a predetermined path and bombards a target. When the bombarding particles interact in the target, a nuclear reaction occurs at a sub-atomic level, resulting in the production of a radioisotope.
Conventional cyclotrons employ a concept called “sector focusing” to constrain the vertical dimension of the accelerated particle beam within the poles of the cyclotron magnet. The magnet poles contain at least two wedge-shaped sectors, commonly known as “hills”, where the magnetic flux is mostly concentrated. The hills are separated by regions, commonly referred to as “valleys”, where the magnet gap is wider. As a consequence of the wider gap the flux density, or field strength, in the valleys is reduced compared to that in the hills.
An exemplary conventional two-pole cyclotron is illustrated in FIG. 1. A conventional two-pole cyclotron has an RF system that includes a plurality of semi-circular or wedge-shaped, hollow electrodes 12a, 12b. The hollow electrodes 12a, 12b, commonly referred to as “dees” because of their shape, each define a curved side 16a, 16b. The dees 12a, 12b are coplanar and are positioned relative to one another such that their respective curved sides 16a, 16b are concentric to define a diameter 20. Each of the dees 12a, 12b defines an entrance 22 to allow access to the interior of the dee and an exit 24. The energy for accelerating the beam 40 of electrically-charged particles is provided by an externally-supplied alternating high voltage. The dees 12a, 12b generally are composed of low-resistance copper so that relatively high traveling currents do not cause uneven voltage distribution within the dee structure.
A cyclotron uses a magnetic field to direct beams of charged particles along a predetermined path. As illustrated in FIG. 1, the two-pole cyclotron includes a magnet system having four magnet poles, each defining a wedge shape. The upper magnet poles 26, 28 protrude downward from the upper magnet yoke 54, toward the lower magnet poles 30, 32 which protrude upward from the lower magnet yoke 56. The magnetic field, which is represented by the arrows 58, is perpendicular to the longitudinal plane of the dees and, therefore, is perpendicular also to the electric field generated by the alternating high voltage. The magnetic field exerts a force that is perpendicular both to the direction of motion of the charged particle and to the magnetic field. Hence, a charged particle in a magnetic field having a constant strength undergoes circular motion if the area defined by the magnetic field is sufficiently large. The diameter of the circular path of the charged particle is dependent on the velocity of the charged particle and on the strength of the magnetic field. It is prudent to note that a magnetic field causes a charged particle to change direction continuously; however, it does not alter the velocity of a charged particle, hence the energy of the charged particle is unaffected.
The cyclotron of FIG. 1 illustrates the general magnetic system. In the limiting case of the “separated sector cyclotron” each hill sector is a complete, separate, stand-alone magnet with its own gap, poles, return/support yoke, and common excitation coil. In this implementation the valleys are merely large void spaces containing no magnet steel. Essentially all the magnetic flux is concentrated in the hills and almost none is in the valleys. In addition to providing tight vertical focusing, the separated-sector configuration allows convenient placement of accelerating electrodes and other apparatus in the large void spaces comprising the valleys.
Vertical focusing of the beam is enhanced by a large ratio of hill field to valley field; the higher the ratio, the stronger are the forces tending to confine the beam close to the median plane. In principle, a tighter confinement, in turn, reduces the required magnet gap without danger of the beam striking the pole faces in the magnet. For a given amount of flux in the gap, a magnet with a small gap requires less electrical power for excitation than does a magnet with a large gap.
Some conventional cyclotrons use electromagnets in the magnetic system. More recently, superconducting magnet technology has been applied to cyclotrons. In superconducting cyclotron designs, the valleys are also large void spaces in which accelerating electrodes and other apparatus may be conveniently emplaced. The magnet excitation for a superconducting cyclotron is usually provided by a single pair of superconducting magnet coils which encircle the hills and valleys. A common return/support yoke surrounds the excitation coil and magnet poles.
FIG. 2 is a representative illustration of a conventional cyclotron focusing on the dees. For simplicity, only two dees 12a, 12b are illustrated. However, there are typically four or more dees used. As will be discussed below, ions are accelerated in a substantially circular, outwardly spiraling path. In devices using fewer dees, either more turns are required, or a higher acceleration voltage is required, or both, in order to energize the ions to the desired level. As The dees 12a, 12b are positioned in the valley of the large electromagnet. Near the center of the dees 12a, 12b is the ion source 81 used for generating charged particles. The ion source 81 is typically an electrical arc device 50 in a gas.
During operation, ions are continuously generated by the ion source 81. A filament located in the ion source assembly creates both negative and positive ions through the addition of electrons or the subtraction of electrons. As the negative ions enter the vacuum tank (not shown) enclosing the dees 12a, 12b, they gain energy due to a high-frequency alternating electric field induced on the dees 12a, 12b. As the negative ions flow from the ion source 81, they are exposed to this electric field as well as a strong magnetic field generated by two magnet poles, one above and one below the vacuum tank. Because these are charged particles in a magnetic field, the negative ions move in a circular path.
When the negative ions reach the edges of the dees 12a, 12b and enter the gap, the RF oscillator changes the polarities on the dees 12a, 12b. The negative ions are repelled as they exit the previously positive but now negatively charged dee 12a, 12b. Each time the particles cross the gap they gain energy, so the orbital radius continuously increases and the particles follow an outwardly spiraling path. The particles are pushed from the first dee 12a and drift along a circular path until they are attracted or pulled by the second dee 12b which has become positively charged. The result is a stream of negative ions which are accelerated in a circular path spiraling outward.
Returning to FIG. 1, all four of the hills 26, 28, 30, 32 and two of the four valleys 34, 36 are visible. The beam 40, during acceleration, is exposed alternately to the strong and weak magnetic fields defined respectively by the hills and valleys along its path to the extraction radius. As the beam 40 passes through each hill region, it bends sharply due to the effect of the strong magnetic field. While in the valley regions, however, the beam trajectory is more nearly a straight path toward the next hill region. This alternating magnetic field provides strong vertical focusing forces to beam particles straying from the median plane during acceleration. These focusing forces direct straying particles back toward the median plane, promoting high beam extraction efficiencies.
As indicated previously, the RF system of a cyclotron supplies an alternating high voltage potential to the dees. In the cyclotron depicted in FIG. 1, each of the two dees 12a, 12b is mounted in a valley region. The beam 40 of positively-charged particles gains energy by being attracted by the dee when the dee has a negative charge, and then by being repelled from the dee as the dee changes to a positive charge. Thus, because a charged particle within the beam 40 passes through both dees 12a, 12b in the course of a single orbit, that charged particle undergoes two increments of acceleration per orbit. Therefore, with every acceleration, the beam 40 of charged particles gains a known, fixed quantity of energy, and its orbital radius increases in predetermined fixed increments until it reaches the extraction radius, which corresponds to the extraction energy of the beam.
The combined effects of the RF system and the magnet system on a charged particle are clarified in the following example: In a positive-ion two-pole cyclotron, such as that depicted in FIG. 1, positively-charged particles in the first dee, which is mounted in the first valley, are accelerated by a negative electric field generated within the first dee. Once these particles exit the first dee and enter the first hill, the magnetic field directs them toward the second dee, which is mounted in the second valley. Upon entering the second dee, the positively-charged particles are accelerated by a negative electric field generated within that dee. Once these particles exit the second dee and enter the second hill, the magnetic field directs them back into the first dee. By repeating this method, the cyclotron predictably and incrementally accelerates the charged particles along a predetermined path, by the end of which the charged particles have acquired their predetermined extraction energy.
As the velocity of a charged particle increases, an ever-strengthening magnetic field is required to maintain the charged particle on the same circular path. Consequently, in a cyclotron, which generates a magnetic field having a constant strength, the incremental acceleration of a charged particle causes the particle to follow an outward, quasi-spiral orbit 70. Thus, the magnetic field is the “bending” force that directs the beam 40 of charged particles along an outward, quasi-spiral orbit 70 around a point centrally located between the dees 12a, 12b. 
Having reviewed the essential principles concerning the functioning of a cyclotron, it is helpful to summarize more of the systems that are included in a cyclotron, all of which are well known in the prior art. The following systems are summarized briefly below: (1) the ion source system, (2) the target system, (3) the shielding system and (4) the radioisotope processing system (optional). Thereafter, the two systems addressed previously in the context of a two-pole cyclotron, i.e., the magnet system and the RF system, are addressed in the context of a four-pole cyclotron.
The ion source system 80 is required for generating the charged particles for acceleration. Although several ion source systems are well known in the prior art, in the interest of brevity, only one of these systems is summarized below. Those skilled in the art will acknowledge that an ion source system comprising an internally, axially-mounted Penning Ion Gauge (PIG) ion source optimized for proton (H+) production is useful for producing fluorine-18, among other positron-emitting radioisotopes. This ion source system ionizes hydrogen gas using a strong electric current. The ionized hydrogen gas forms plasma, from which protons (H+ ions) are extracted for acceleration using a bias voltage.
After the beam 40 of charged particles acquires its extraction energy, it is directed into the target system 88. Target systems are well known in the prior art. In general, the beam exits the magnetic field 58 at the predetermined location 90 and enters the accelerator beam tube 92, which is aligned with the target entrance 94. A collimater 96, which consists of a carbon disk defining a central hole, is mounted at the target entrance 94, and as the beam 40 passes through the collimater 96, the collimater 96 refines the profile of the beam. The beam 40 then passes through the target window 98, which consists of an extremely thin sheet of foil made of a high-strength, non-magnetic material such as titanium. Thereafter, the beam 40 encounters the target substance 100, which is positioned behind the target window 98. The beam 40 bombards the target substance 100, which may comprise a gas, liquid, or solid, generating the desired radioisotope through a nuclear reaction.
Cyclotrons vary in the method used to extract the beam such that it exits the magnetic field at the predetermined location. Regarding a negative-ion cyclotron (not shown), the beam, which initially consists of negatively-charged particles, is extracted by changing its polarity. A thin sheet of carbon foil is positioned in the path of the beam, specifically, along the extraction radius. As the beam interacts with the carbon foil, the negatively-charged particles lose their electrons and, accordingly, become positively charged. As a result of this change in polarity, the magnetic field forces the beam, now consisting of positively-charged particles, in the opposite direction instead, causing the beam to exit at the predetermined location and enter the accelerator beam tube. It is important to note that the carbon foil acquires only a trivial amount of radioactivity as a result of its interaction with the beam. Regarding a positive-ion cyclotron, however, carbon foil cannot be used to change the polarity of the beam because the beam initially consists of positively-charged particles, which already have an electron deficit. Instead, as depicted in FIG. 1, a conventional positive-ion cyclotron uses a magnet extraction mechanism that includes two blocks 102, 104 made of a metal such as nickel. The first block 102 is affixed to an upper magnet pole such that it protrudes downward toward a lower magnet pole. The second block 104 is affixed, opposite the first block, to a lower magnet pole such that it protrudes upward toward an upper magnet pole. The blocks 102, 104 are positioned above and below the extraction radius, respectively, and they operate to perturb the magnetic field such that its effect on the beam, as it passes between the blocks 102, 104, is mitigated at that location. Hence, the “bending” force exerted by the magnetic field on the beam at that location is weakened, causing the beam to exit at the predetermined location and enter the accelerator beam tube. Inevitably, the edges of the beam interact with the two blocks 102, 104, converting them, at least in part, into a metal radioisotope that has a long half-life. Due to this long half-life, the metal radioisotope accumulates in the blocks 102, 104 during operation, rapidly becoming a significant, enduring, and worrisome source of harmful radiation. In sum, in comparison to a negative-ion cyclotron, a conventional positive-ion cyclotron is disadvantaged in that its magnet extraction mechanism is a major source of harmful radiation.
Harmful radiation is generated as a result of operating a cyclotron, including a negative-ion cyclotron, and it is attenuated to acceptable levels by a shielding system, several variants of which are well known in the prior art. A cyclotron has several sources of radiation that warrant review. First, prompt high-energy gamma radiation and neutron radiation, a byproduct of nuclear reactions that produce radioisotopes, are emitted when the beam, or a particle thereof, is deflected during acceleration by an extraction mechanism into an interior surface of the cyclotron. As stated previously, such deflections are a major source of harmful radiation in a conventional positive-ion cyclotron. In the target system 88, prompt high-energy gamma radiation and neutron radiation are generated by the nuclear reaction that occurs as the beam 40 bombards the target substance 100, producing the desired radioisotope. Also in the target system 88, induced high-energy gamma radiation is generated by the direct bombardment of target system components such as the collimater 96 and the target window 98. Finally, residual radiation is indirectly generated by the nuclear reaction that yields the radioisotope. During the nuclear reaction, neutrons are ejected from the target substance 100, and when they strike an interior surface of the cyclotron, gamma radiation is generated. Although commonly composed of layers of exotic and costly materials, shielding systems only can attenuate radiation; they cannot absorb all of the gamma radiation or other ionizing radiation.
Following the generation of the desired radioisotope, the target substance 100 commonly is transferred to a radioisotope processing system. Such radioisotope processing systems are numerous and varied and are well known in the prior art. A radioisotope processing system processes the radioisotope primarily for the purpose of preparing the radioisotope for the tagging or labeling of molecules of interest, thereby enhancing the efficiency and yield of downstream chemical processes. For example, undesirable molecules, such as excess water or metals, are extracted.
FIG. 3 depicts some of the components of the magnet system 120 and the RF system 150 typical of a positive-ion four-pole cyclotron. The magnet system 120 comprises eight magnet poles, each defining a wedge shape. Four of the magnet poles extend from the upper magnet yoke downward, toward the remaining four magnet poles, which extend upward from the lower magnet yoke. As stated previously, magnet poles are often called “hills,” and the hills define recesses that are often called “valleys.” In FIG. 3, only seven of the hills 122, 124, 126, 128, 130, 132, 133 and six of the valley regions 134, 136, 138, 120, 122, 124 are at least partially depicted. The beam 40, during acceleration, is exposed alternately to the strong and weak magnetic fields defined respectively by the hills and valleys along its path to the extraction radius. The RF system 150 of a four-pole cyclotron includes four dees 152, 154, 156, 158, each having a wedge shape. Each of the four dees 152, 154, 156, 158 is mounted in a valley region 134, 136, 138, 120. The beam 40 of charged particles gains energy by being attracted to, and then repelled from, each dee through which it passes. Thus, because a charged particle within the beam 40 passes through all four dees 152, 154, 156, 158 in the course of a single orbit, that charged particle, which experiences an increment of acceleration per dee, undergoes four increments of acceleration per orbit.
Cyclotrons that are typical of the art are those devices disclosed in the following U.S. Pat. Nos.:
Patent No.Inventor(s)Issue Date1,948,384E. O. LawrenceFeb. 20, 19344,206,383V. G. Anicich et al.Jun. 3, 19804,639,348W. S. JarnaginJan. 27, 19875,463,291L. Carroll et al.Oct. 31, 19955,818,170T. Kikunaga et al.Oct. 6, 19986,060,833J. E. VelazcoMay 9, 20006,163,006F. C. Doughty et al.Dec. 19, 20006,396,024F. C. Doughty et al.May 28, 20026,523,338G. Kornfeld et al.Feb. 25, 20032004/0046116J. B. Schroeder et al.Mar. 11, 20042006/0049902L. KaufmanMar. 9, 2006
Of these patents, Lawrence, in his '384 patent, discloses a method and apparatus for the acceleration of ions. The Lawrence patent is based primarily upon the cumulative action of a succession of accelerating impulses, each requiring only a moderate voltage, but eventually resulting in an ion speed corresponding to a much higher voltage. According to Lawrence, this is accomplished by causing ions or electrically charged particles to pass repeatedly through accelerating electric fields in such a manner that the motion of the ion or charged particle is in resonance or synchronism with oscillations in the electric accelerating field or fields.
Anicich et al., in their '383 patent, disclose a miniaturized ion source device in an air gap of a small permanent magnet with a substantially uniform field in the air gap of about 0.5 inch. The device and permanent magnet are placed in an enclosure which is maintained at a high vacuum (typically 10−7 torr) into which a sample gas can be introduced. The ion-beam end of the device is placed very close to an aperture through which an ion beam can exit into apparatus for an experiment.
Jarnagin, in his '348 patent, discloses a re-circulating plasma fusion system. The '348 patent claims to include a plurality of recyclotrons, each comprising cyclotron means for receiving and accelerating charged particles in spiral and work conservative pathways, and output means for forming a beam from particles received. The cyclotron means used by Jarnagin includes a channel shaped electromagnet having a pair of indented polefaces oriented along an input axis and defining an input magnetic well. The cyclotron further includes a pair of elongated linear electrodes centered along the input magnetic well arranged generally parallel to the input axis and having a gap therebetween. A tuned oscillator means is connected to the electrodes for applying an oscillating electric potential thereto. The output means includes an inverter means including an electromagnet having a polarity opposite that of the channel shaped electromagnet oriented contigously therealong for extracting fully accelerated particles from the cyclotron means. A reinverter means includes an electromagnet having a polarity the same as that of the channel shaped electromagnet for correcting the flight path of the extracted particles, the inverter means and the reinverter means defining an output axis, along which the output means directs the beam. The recyclotrons are arranged so that particles of the output beam are received by the input magnetic well of an opposing similar recyclotron.
Carroll, et al., in their '291 patent, disclose a cyclotron and associated magnet coil and coil fabricating process. The cyclotron includes a return yoke defining a cavity therein. A plurality of wedge-shaped regions called “hills” are disposed in the return yoke, and voids called “valleys” are defined between the hills. A single, substantially circular magnet coil surrounds and axially spans the hills and the valleys.
In the '170 patent, Kikunaga et al., disclose a gyrotron system including an electron gun that produces an electron beam. A magnetic field generating unit comprises a permanent magnet and two electromagnets, and is capable of generating an axial magnetic field that drives electrons emitted from the electron gun for revolving motion. A cavity resonator causes cyclotron resonance maser interaction between the revolving electrons and a high-frequency electromagnetic field resonating in a natural mode. A collector collects the electron beam that has traveled through the cavity resonator. An output window is provided, through which a high-frequency wave produced by the cyclotron resonance maser interaction propagates.
Velazco, in the '833 patent, discloses an electron beam accelerator utilizing a single microwave resonator holding a transverse-magnetic circularly polarized electromagnetic mode and a charged-particle beam immersed in an axial focusing magnetic field.
In their '006 patent, Doughty et al., disclose a plasma-producing device wherein an optimized magnet field for electron cyclotron resonance plasma generation is provided by a shaped pole piece.
In their '024 patent, Doughty et al., disclose a method and apparatus for integrating multipolar confinement with permanent magnetic electron cyclotron resonance plasma sources to produce highly uniform plasma processing for use in semiconductor fabrication and related fields. The plasma processing apparatus includes a vacuum chamber, a workpiece stage within the chamber, a permanent magnet electron cyclotron resonance plasma source directed at said chamber, and a system of permanent magnets for plasma confinement about the periphery of the chamber.
Kornfeld et al., in the '338 patent, disclose a plasma accelerator arrangement in particular for use as an ion thruster in a spacecraft. A structure is proposed in connection with which an accelerated electron beam is admitted into an ionization chamber with fuel gas, and is guided through the ionization chamber in the form of a focused beam against an electric deceleration field, said electric deceleration field acting at the same time as an acceleration field for the fuel ions produced by ionization.
In Published Application No. 2004/0046116, Schroeder et al., disclose a negative ion source placed inside a negatively-charged high voltage terminal for emitting a beam which is accelerated to moderate energy and filtered by a momentum analyzer to remove unwanted ions. Reference ions such as carbon-12a are deflected and measured in an off-axis Faraday cup. Ions of interest, such as carbon ions of mass 12b, are accelerated through 300 kV to ground potential and passed through a gas stripper where the ions undergo charge exchange and molecular destruction. The desired isotope, carbon-12b along with fragments of the interfering molecular ions, emerges from the stripper into a momentum analyzer which removes undesirable isotope ions. The ions are further filtered by passing through an electrostatic spherical analyzer to remove ions which have undergone charge exchange. The ions remaining after the spherical analyzer are transmitted to a detector and counted.
In Published Application No. 2006/0049902, Kaufman defines a plurality of permanent magnets to enhance radiation dose delivery of a high energy particle beam. The direction of the magnetic field from the permanent magnets may be changed by moving the permanent magnets.
A cyclotron (or other particle accelerator), although required for the production of positron radiopharmaceuticals, was (and still is) uncommon due to its high price, high cost of operation, and stringent infrastructure requirements relating to it immensity, weightiness and high energy consumption. Consequently, at one time, a great majority of institutions did not have a PET scanner. Thereafter, however, some businesses, e.g., CTI PETNet, established relatively efficient distribution networks to supply hospitals and imaging centers with positron radiopharmaceuticals, thereby allowing them to avoid the substantial costs and other impracticalities associated with cyclotrons. Consequently, the number of PET scanners in operation increased dramatically relative to the number of cyclotrons in operation. However, because the half-lives of positron radiopharmaceuticals are short, there still exists an inherent inefficiency in a radiopharmaceutical distribution network that cannot be overcome. This inefficiency results, in part, from the radioactive decay of the radiopharmaceutical during transport from the site of production to the hospital or imaging center. It results also, in part, from the limitations inherent in the conventional (macroscale) chemical apparatuses that receive the radioisotopes and use them in synthesizing radiopharmaceuticals. The processing times that such apparatuses require are lengthy relative to the half-lives of most clinically-important positron-emitting radioisotopes. For example, CTI's Explora FDG4, an efficient macroscale chemical apparatus, requires forty-five (45) minutes to convert nucleophilic fluorine-18 ([18F]F−) into [18F]fluorodeoxyglucose ([18F]FDG), a glucose analogue that is commonly used in PET. Fluorine-18 has a half-life of only 110 minutes. Also, to generate the relatively large quantities of [18F]F− required of the Explora FDG4, which is on the order of curies (Ci), the bombardment of the target material generally continues for approximately two (2) hours. During that time, however, a significant percentage of the newly generated [18F]F− decays back to its original oxygen state. Also, the percent yield of the macroscale chemical apparatus is only approximately 50 to 60%. The limitations of macroscale chemical apparatuses are even more evident when preparing biomarkers that are labeled with positron-emitting radioisotopes having even shorter half-lives, such as carbon-11 (t1/2=20 min), nitrogen-13 (t1/2=10 min), and oxygen-15 (t1/2=2 min).
In recent years, however, a promising new discipline, sometimes referred to as microreaction technology, has emerged. A microreactor is a miniaturized reaction system fabricated, at least in part, using methods of microtechnology and precision engineering. The first prototype microreactors for chemical processes, including chemical synthesis, were manufactured and tested in the early 1990's. The characteristic linear dimensions of the internal structures of a microreactor, such as fluid channels, generally are in the nanometer to millimeter range. For example, the fluid channels in a microreactor typically have a diameter of between approximately a few nanometers and approximately a few millimeters. The length of such channels, however, can vary significantly, i.e., from on the order of millimeters to on the order of meters, depending on the function of the channel. There are exceptions, however, and microreactors having characteristic linear dimensions that are shorter or longer have been developed. A microreactor may include only one functional component, and that component may be limited to a single operation, such as mixing, heat exchange, or separation. Examples of such functional components include micropumps, micromixers, and micro heat exchangers. As more than one operation generally is necessary to perform even the simplest chemical process, more complex systems, sometimes referred to as integrated microreaction systems, have been developed. Typically, such a system includes at least several different functional components, and the configuration of such systems can vary significantly depending on the chemical process that the system is engineered to perform. Additionally, integrated microreaction systems that include arrays of microreactors have been developed to provide continuous-flow production of chemicals.
In microreaction systems, an increase in throughput is achieved by increasing the number of microreactors (numbering up), rather than by increasing the dimensions of the microreactor (scaling up). Thus, additional microreactors are configured in parallel to achieve the desired increase in throughput. Numbering up is the preferred method because only it can preserve the advantages unique to a microreaction system, which are summarized below and are derived from the minuscule linear dimensions of the system's internal structures.
First, as the linear dimensions of a reactor decrease, the surface area to volume ratio of the reactor increases. Accordingly, the surface area to volume ratio of the internal structures of a microreactor generally range from 10,000 to 50,000 m2/m3, whereas typical laboratory and production vessels usually do not exceed 1000 m2/m3 and 100 m2/m3, respectively. Because of its high surface area to volume ratio, a microreactor has an exchange surface for heat transfer and mass transport that is relatively far greater than that of a conventional reactor. This promotes very rapid heating, cooling, and mixing of reagents, which can improve yields and decrease reaction times. This is especially significant because, when synthesizing fine chemicals (e.g., radiopharmaceuticals) using conventional systems, the reaction time usually is extended beyond what is kinetically necessary to compensate for the relatively slow heat transfer and mass transport typical of a system having a conventional surface area to volume ratio. When using a microreaction system, the reaction time does not need to be extended significantly to allow for effective heat transfer and mass transport. Consequently, chemical synthesis is significantly more rapid, and the percent yield of a microreaction system is significantly higher, especially in comparison to a conventional (macroscale) system using a batch-production process.
Second, it is critical to note that the behavior of a fluid, namely a liquid or a gas, in a milliscale, microscale, or nanoscale system differs significantly from the behavior of a fluid in a conventional (macroscale) system. In a system that is not at equilibrium regarding one or more physical properties (e.g., concentration, temperature, or pressure), the linear dimensions of the system are factors in determining the gradient relating to each physical property. As linear dimensions decrease, each gradient increases, thereby increasing the force driving the system toward equilibrium. For example, in the absence of mixing, molecules of a gas spontaneously undergo random movement, the result of which is the net transport of those molecules from a region of higher concentration to one of lower concentration, as described in Fick's laws of diffusion. More particularly, Fick's first law of diffusion states that the flux of the diffusing material in any part of the system is proportional to the local concentration gradient. Thus, in a system having linear dimensions on the order of nanometers, for example, the diffusional flux would very rapidly drive the system to constant concentration. To explain further using another method, the mobility of water can be expressed in terms of a diffusion coefficient, D, which for water equals approximately 2.4×10−5 cm2/s at 25° C., where D is a proportionality constant that relates the flux of amount of entities to their concentration gradient. The average distance s traversed in time t depends on D, according to the expression: s=(4Dt)1/2. Thus, a single water molecule diffuses an average distance of 98 micrometers per second at 25° C. This rate discloses that a water molecule in a water solution can traverse a channel or reaction chamber having a diameter of 100 micrometers extremely quickly, i.e., in approximately 1.0 second. In a microreaction system, the average distance s is extremely long relative to the dimensions of the internal structures of the system. Accordingly, diffusion is dominant, and profiles of concentration are essentially linear and time-independent. Similar principles apply in chemical diffusion, which is the diffusion under the influence of a gradient in chemical composition. In other words, in a microreaction system, the force driving the interdiffusion of two or more miscible reagents nearly instantaneously eliminates any concentration gradients. Similarly, gradients relating to other physical properties, including temperature and pressure, are nearly instantaneously eliminated. A microreaction system, therefore, can equilibrate nearly instantaneously both thermally and compositionally. Accordingly, such a system is highly responsive and allows for very precise control of reaction conditions, improving reaction kinetics and reaction product selectivity. Such a system allows also for a high degree of repeatability and process optimization. These factors in combination significantly improve yields and reduce processing times.
Third, a microreaction system may also alter chemical behavior for the purpose of enhancing performance. Some microreaction systems include extremely minuscule reaction vessels, cavities, or clefts that can partially encapsulate molecules of a reagent, thereby providing an environment in which interaction via molecular forces can modify the electronic structure of reagent molecules. Steric interactions are possible also, including those that influence the conformation of a reagent molecule or those that affect the free rotation of a chemical group included in a reagent molecule. Such interactions modify the reactivity of the reagents and can actively change the chemistry underlying the chemical process by altering the mechanism of the reaction.
Other advantages of using a microreaction system, instead of a conventional (macroscale) system, include increased portability, decreased reagent consumption, and decreased hazardous waste generation. In sum, microreaction systems, due at least in part to their small size and efficiency, facilitate the synthesis of fine chemicals at, or proximate to, the site of consumption. Such systems are capable of providing on-site and on-demand synthesis of fine chemicals, including radiopharmaceuticals.
More recently, in 2002, a scientific article disclosed the development of “high-density microfluidic chips that contain plumbing networks with thousands of micromechanical valves and hundreds of individually addressable reaction chambers.” T. Thorsen, S. J. Maerkl, S. R. Quake, Microfluidic Large-Scale Integration, Science, Vol. 298, no. 5593 (Oct. 18, 2002) pp. 580-584. The article disclosed also that “[t]hese fluidic devices are analogous to electronic integrated circuits fabricated using large-scale integration.” Not surprisingly, at least one manufacturer of high-density microfluidic chips (Fluidigm Corporation) refers to them as integrated fluidic circuits (IFCs). The term microfluidics generally is used broadly to refer to the study of fluid behavior in microscale, nanoscale, or even picoscale systems. As is common in the terminology of emerging scientific or engineering disciplines, there is no unanimity on a definition of microfluidics, and there likely is at least some overlap between microfluidics and the discipline of microreaction technology described previously. Generally, a microfluidic system is distinguishable in that it processes fluids on a chip that defines a fluidic circuit, where the chip is under digital control and the fluid processing is performed using the fluidic circuit, which includes at least one reaction channel, chamber, compartment, reservoir, vessel, or cleft having at least one cross-sectional dimension (e.g., diameter, depth, length, width, height) on the order of micrometers, nanometers, or even picometers for altering fluid behavior and, possibly, chemical behavior for the purpose of enhancing performance. Accordingly, a microfluidic system enjoys the advantages inherent in a microreaction system that were set forth previously. At least some microfluidic systems can be thought of as including a fluidic chip that incorporates a microreactor. Microfluidic systems are able to exercise digital control over, among other things, the duration of the various stages of a chemical process, leading to a well-defined and narrow distribution of residence times. Such control also enables extremely precise control over flow patterns within the system. Thus, within a single microfluidic chip, especially one with integrated microvalves, the automation of multiple, parallel, and/or sequential chemical processes is possible. Microfluidic chips generally are manufactured at least in part using lithography (e.g., photolithography, multi-layer soft lithography).
In 2005, a scientific article disclosed the development of “a microfluidic chemical reaction circuit capable of executing the five chemical processes of the syntheses of both [18F]FDG and [19F]FDG within a nanoliter-scale reaction vessel.” C.-C. Lee, et al., Multistep Synthesis of a Radiolabeled Imaging Probe Using Integrated Microfluidics, Science, Vol. 310, no. 5755, (Dec. 16, 2005), pp. 1793-1796. Specifically, the article stated that “[t]he production of [18F]FDG [was] based on five sequential chemical processes: (i) concentration of the dilute [18F]fluoride mixture solution (<1 ppm, specific activity ˜5000 to 10,000 Ci/mmol), obtained from the proton bombardment of [18O]water at a cyclotron facility; (ii) solvent exchange from water to acetonitrile (MeCN); (iii) [18F]fluoride substitution of the triflate group in the D-mannose triflate precursor in dry MeCN; (iv) solvent exchange from MeCN to water; and (v) acidic hydrolysis of the fluorinate intermediate to obtain [18F]FDG.” Regarding step (i), the article stated further that “an in situ ion-exchange column was combined with a rotary pump to concentrate radioisotopes by nearly three orders of magnitude, thereby optimizing the kinetics of the desired reactions.” Beyond the five sequential chemical processes, the article disclosed that the microfluidic chip incorporated “digital control of sequential chemical steps, variable chemical environments, and variable physical conditions” and had “the capability of synthesizing the equivalent of a single mouse dose of [18F]FDG on demand.” The chip also “accelerate[d] the synthetic process and reduce[d] the quantity of reagents and solvents required.” The article disclosed further that “[t]his integrated microfluidic chip platform can be extended to other radiolabeled imaging probes.” Moreover, the article disclosed “a second-generation chemical reaction circuit with the capacity to synthesize larger [18F]FDG doses” that “should ultimately yield large enough quantities (i.e., >100 mCi) of [18F]FDG for multiple human PET scans, which typically use 10 mCi per patient.”
Additionally, Nanotek, LLC, a company based in Walland, Tenn., manufactures and distributes a microfluidic device called the MinuteManLF. This commercially-available state-of-the-art microfluidic device can synthesize [18F]FDG in as little as 100 seconds, while obtaining percent yields as high as 98%. Additionally, the MinuteManLF can be used to synthesize [18F]fluoro-3′-deoxy-3′-L-fluorothymidine ([18F]FLT), a PET biomarker that is particularly useful for monitoring tumor growth and response by enabling in vivo quantitative imaging of cellular proliferation.